The Internet of Things and Ubiquitous Intelligence (4)

Editor's Desk:
    The traditional Internet is oriented towards person-to-person connection, whereas the Internet of things (IoT) is oriented towards connections between inanimate objects. IoT covers a larger range of connections and involves more semantics than traditional Internet. Traditional Internet and telecom networks focus on information transfer, but IoT focuses on information services. By combining sensor networks, Internet, telecom networks, and cloud computing platform, IoT can sense, recognize, affect, and control the physical world. The physical world can be unified with the virtual world and human perception. In this part, we discuss cloud computing and the cyber-physical system (CPS).

8 Cloud Computing
    A distributed CPU is the core component in cloud computing and plays a pivotal role in the development of the Internet of things. Combining cloud computing and the IoT can boost the entire industry and value chain if capability and resource sharing, quick service deployment, expansion of new human-thing interactive services, and deep data mining are optimized. When the Internet of things reaches a certain level, its dependence on cloud computing will be greater.

8.1 Concept and Current Status
    With the development of electronics, communications, computers, and network technologies, cloud computing is an inevitable stage in the evolution of turing computing to network computing. Based on the Internet, network computing provides applications with hardware, infrastructure, platform, software, and storage services. The term “cloud” vividly describes the characteristics of cloud computing: virtualization, transparency, scalability, and elasticity. It is an apt term for describing the abstraction of underlying infrastructure.
Cloud computing integrates distributed computing, parallel computing and grid computing. At its core, it virtualizes computing, resources of large data centers and provides these to users as a service.

8.2 Key Technologies
    Cloud computing is a form of data-intensive super computing. It has unique data storage and management technologies and a unique programming model.

8.2.1 Virtualization
    Virtualization is at the core of cloud computing, and is the basis for all cloud services applications. It expands hardware capacity, makes software reconfiguration reduces easier, overheads of software virtual machines is reduced, and supports multiple operating systems. With virtualization, use of the server is greatly enhanced.

8.2.2 Data Storage and Management
    In cloud computing, data is highly reliable and highly available because of storage redundancy. Data storage technologies involve cluster computing, data redundancy, and distributed storage.

    The frequency of data access is much higher than that of data update because of one characteristic of cloud computing—analyzing the stored and accessing massive data [1]. Therefore, cloud data management focuses on optimizing the data access operation.

8.2.3 Parallel Programming
    The cloud computing programming model must be very simple, and  complicated parallel execution and task scheduling must be hidden in the background. The programming model must be transparent, enabling users to develop programs for specific purposes.

    Most cloud computing systems adopt Google’s MapReduce [2]. This is a distributed parallel programming model and also an efficient task-scheduling model.

8.3 Combining with the Internet of Things
    Cloud computing and IoT are complementary. Development of IoT requires the powerful processing and storage capabilities of cloud computing. If cloud computing infrastructure is used to mine, process, and analyze mass data collected on the ubiquitous sensing layer, IoT can quickly, accurately, and intelligently manage and control the physical world and provide technical support for ubiquitous services. The IoT will also be the largest user of cloud computing services, which will, in turn, lay the foundation for more successful cloud computing.

9 Cyber-Physical System
    Cyber-physical system (CPS) was first proposed by the US National Science Foundation (NSF) in 2006. CPS is expected to become the third wave of information technology following the computer and the Internet. CPS ubiquitous sensing, control, and computing and the close integration of human, machines, and things are the ultimate goals of IoT.

9.1 CPS Overview
    CPS is complicated a multidimensional, embedded system that is integrated with computing, network and physical environments. With the integration of 3C (computation, communication, and control) technologies, CPS delivers real-time sensing, dynamic control, and information services for large engineering systems. Through interactive, cyclic feedback in computing and physical processes, CPS closely integrates these processes and adds or expands new functions in real time. In this way, a physical entity can be monitored and controlled in a secure, reliable, efficient, and real-time manner [5].

9.2 Characteristics
    CPS has the following characteristics:

• It integrates computing, communication, and control because it embeds computing capability deeply into each physical subsystem. It aims for precise control over the physical processes in networks.
• It requires the integration of computing and control technologies. To connect the cyber world to the physical world, CPS should integrate computing technologies (which are discrete event-relevant and indifferent to time and space) with the control technologies (which are continuous process-relevant and time space-focused), enabling the discrete computing process to interact and tightly couple with the continuous physical process.
• It is adaptive to dynamic changes in the physical world; it has powerful reorganization and reconstruction capabilities; it can esaily; and it is easy to connect with other CPS subsystems.
• Because it has multidimensional time-space complexity, CPS should be an open, trustable, and predictable embedded system. The embedded computing system of CPS interconnects and interoperates with other cyber systems via the Internet. Furthermore, CPS has stepped into fields that are closely related to national infrastructure and people’s daily lives. It is sensitive to security, and its technologies and products must be precise and highly reliable.

9.3 Challenges
    To realize the goals of CPS, there are some technical challenges to be addressed. A good architecture is critical in maintaining scalability, durability, diversity and continuous technical innovation in CPS [3]. It is also the key to customer investment.

    To ensure sustainable development of CPS, the architecture should be stable, a clear roadmap should be devised that can lead to a targeted study based on core CPS architecture, and a development schedule should be in place so that technical achievements are planned and predicted. The natural development cycle should be balanced.

    There are several challenges in the development of CPS. The theoretical basis differs from the application model. The key to the success of a technology is its deployment in actual environments. It is also necessary to maintain the integrity, reliability and security of the architecture. Because of uncertainty in the external environment and potential changes, CPS should respond to a system failure or malicious attack automatically, autonomously, and quickly. According to the characterisics described in 9.2 the OS and architecture should be able to manage redundant resources on the interconnected device layers, monitor errors in user applications and physical components, and quickly recover CPS from a system failure. Electricity CPS is a critical infrastructure, and guaranteeing its security is an important issue.

10 Summary
    The IoT, cloud computing, and CPS interact and couple with each other. IoT is the intuitive application of CPS, and cloud computing provides technical support for IoT information service. The demands of economy and society on the Internet of things and CPS go far beyond existing applications. Therefore, from the perspective of economic development and technical innovation, IoT and CPS are a trend in the development of global information technologies and industry. They will have a profound effect on existing industrial structure, become the core competence in the new industrial structure, and cultivate new economic growth.
References
[1] Q.Chen and Qianni Deng, “Cloud computing and its key technologies,” in Journal of Computer Applications, vol.29,no.9,pp.2562-2567,2009.
[2] J.Dean and S. Ghemawat, “MapReduce:Simplied Data Processing on Large Clusters,”in Proc. 6th USENIX Symp. on Operation Syst. Design and Implementation (OSDI'04), New York,2004,pp.137150.
[3] T.Aldridge, G.Allee and A.Gorius, “Syst. Architecture and Industry Structure as Interrelated Foundations for Progress in Cyber-physical Energy Syst.,” in Proc.National Workshop on Research Directions for Future Cyber-physical Energy Syst., Baltimore,2009.